International Long Distance Calling: Don't Assume, AssessInternational Long Distance Calling: Don't Assume, Assess

IT decision makers in provider organizations should verify three factors are part of an ILD platform’s policies and procedures: reliability, redundancy and recovery.

The United States is an increasingly mobile society, with strong global connections. It's estimated that one out of every four adults living in the U.S. currently make international calls, and that number will undoubtedly continue to rise. Providing international long distance calling products can add significant additional revenues -- millions of dollars even -- each year for mobile operators, prepaid distributors, cable companies and agents.

The subscriber numbers and potential ROI are compelling. More than 40 million people currently residing within the U.S. (roughly 13% of the total population) are not U.S. citizens by birth, according to the U.S. Census Bureau Survey, 2010. Along with expatriates, military personnel, and tourists, many of these subscribers need international phone service, and research tells us that they would like to use the convenience of their mobile phone to place their calls.

By taking advantage of turnkey cloud-based international long distance or ILD platforms, these calls can be affordable for subscribers and a profitable revenue streams for providers. However, as many expats can attest, if call quality and consistency is not acceptable, another method of making the call will quickly be found.

Although providers may be eager to capitalize on the lucrative ILD market, the IT decision makers in the provider organization cannot simply assume that the vendor providing the ILD telephony service has taken the necessary steps to ensure that in the event of a catastrophe, several minutes of service outage won't stretch into hours. After all, they claim: "We take care of everything -- from infrastructure to switches to staff." But do they?

It's critical for providers that offer ILD calling to have a comprehensive disaster recovery plan, with 24/7 operational monitoring and recovery safeguard in place. When it comes to international calling, do not assume, assess! When making recommendations, IT decision makers in provider organizations should verify that these factors are part of an ILD platform's policies and procedures, commonly referred to as the three R's: reliability, redundancy and recovery.

When a consumer wants to make an international call, they naturally expect their call to go through as easily as one placed in-country -- regardless of when the call takes place. Holidays generate predictable increases in call volumes that can be planned for well in advance. Unfortunately, unplanned spikes do occur. Consider that when significant weather events, such as earthquakes or flooding occurs, consumers naturally reach out to friends, family and businesses. A well-designed ILD system should be able to withstand unexpected high call volumes so that the call is not affected by network congestion and delays.

For example, if a normal traffic route is compromised by, say a blizzard, the system should automatically select alternate routes within a region that can also reach an affected area.

An ILD system should also have the ability to scale quickly. Based on load testing, what is the maximum amount of traffic the system can handle? How many calls per second? Can the system handle a load spike of 100 % without any negative impact? A system may be called upon to handle a 200 or 300 % spike in call volume on a given day. The use of an "N+1" load sharing architectural approach on all critical systems provides the ability to expand to the needed capacity quickly.

The actual call quality the subscriber experiences should also be monitored and analyzed in real-time for fast-busy signals, dead air, poor audio quality or cut-offs down to a specific geographical area. If a route does not meet minimum quality standards, traffic should be re-routed and adjustments made.

A marginally reliable platform and undependable service will undoubtedly lead to consumers searching for a different way to complete their international calls. Therefore, a hosted-services platform should ensure all pieces of the architecture are fully redundant, with no single points of failure, to maximize uptime and provide the best level of service to end users.

If a server fails, there must be a hot standby in place that can seamlessly take over to avoid an outage or loss of functionality. Not only should each disk drive be mirrored and hot spares available to automatically take over while a failed drive is replaced, but each disk enclosure that contains these drives should also be mirrored to guard against a total disk enclosure failure. Redundant SAN connections from each server to each disk enclosure must be in place to prevent a networking failure from causing loss of service.

Additionally, each piece of the call processing architecture must be redundant to allow for deployment of new code with minimal to no downtime, as well as to protect against coding errors that can cause an application to fail.

The network operation center personnel must monitor the health of the call processing network 24/7, and the hosted-services platform should automatically and continuously monitor every aspect of the platform and sound an alarm if a failure occurs.

When international calling is involved, it's a given that the network operation center must actively monitor global events, as well as have alternate locations in place for network operations and offsite data centers.

An offsite center located in a different geographical area should be available in the event that infrastructure is impacted throughout a region by large-scale failure. Operational procedures should be in place to ensure critical systems can be recovered to continue delivery of core services.

Cloud-based turnkey international long distance calling products and services at competitive rates can be an extremely profitable business strategy for MVNOs and others. However, that requires trusting infrastructure, switches and technical staff to an outside vendor. When a disruption of service could potentially affect the level of quality consumers expect, it is incumbent upon those with IT expertise to ensure their provider is making the best choice and verify that the proper recovery protocols and procedures are in place. Don't assume, assess.

Karthik Srinivasan is Director of Systems Engineering at NetworkIP.